T^(h#dZddlmZddlmZddlmZddlmZddl m Z e je Z Gdd eZGd d eZd d gZy ) zX-MOD configuration) OrderedDict)Mapping)PretrainedConfig) OnnxConfig)loggingcZeZdZdZdZ dfd ZxZS) XmodConfigan This is the configuration class to store the configuration of a [`XmodModel`]. It is used to instantiate an X-MOD model according to the specified arguments, defining the model architecture. Instantiating a configuration with the defaults will yield a similar configuration to that of the [facebook/xmod-base](https://huggingface.co/facebook/xmod-base) architecture. Configuration objects inherit from [`PretrainedConfig`] and can be used to control the model outputs. Read the documentation from [`PretrainedConfig`] for more information. Args: vocab_size (`int`, *optional*, defaults to 30522): Vocabulary size of the X-MOD model. Defines the number of different tokens that can be represented by the `inputs_ids` passed when calling [`XmodModel`]. hidden_size (`int`, *optional*, defaults to 768): Dimensionality of the encoder layers and the pooler layer. num_hidden_layers (`int`, *optional*, defaults to 12): Number of hidden layers in the Transformer encoder. num_attention_heads (`int`, *optional*, defaults to 12): Number of attention heads for each attention layer in the Transformer encoder. intermediate_size (`int`, *optional*, defaults to 3072): Dimensionality of the "intermediate" (often named feed-forward) layer in the Transformer encoder. hidden_act (`str` or `Callable`, *optional*, defaults to `"gelu"`): The non-linear activation function (function or string) in the encoder and pooler. If string, `"gelu"`, `"relu"`, `"silu"` and `"gelu_new"` are supported. hidden_dropout_prob (`float`, *optional*, defaults to 0.1): The dropout probability for all fully connected layers in the embeddings, encoder, and pooler. attention_probs_dropout_prob (`float`, *optional*, defaults to 0.1): The dropout ratio for the attention probabilities. max_position_embeddings (`int`, *optional*, defaults to 512): The maximum sequence length that this model might ever be used with. Typically set this to something large just in case (e.g., 512 or 1024 or 2048). type_vocab_size (`int`, *optional*, defaults to 2): The vocabulary size of the `token_type_ids` passed when calling [`XmodModel`]. initializer_range (`float`, *optional*, defaults to 0.02): The standard deviation of the truncated_normal_initializer for initializing all weight matrices. layer_norm_eps (`float`, *optional*, defaults to 1e-12): The epsilon used by the layer normalization layers. position_embedding_type (`str`, *optional*, defaults to `"absolute"`): Type of position embedding. Choose one of `"absolute"`, `"relative_key"`, `"relative_key_query"`. For positional embeddings use `"absolute"`. For more information on `"relative_key"`, please refer to [Self-Attention with Relative Position Representations (Shaw et al.)](https://arxiv.org/abs/1803.02155). For more information on `"relative_key_query"`, please refer to *Method 4* in [Improve Transformer Models with Better Relative Position Embeddings (Huang et al.)](https://arxiv.org/abs/2009.13658). is_decoder (`bool`, *optional*, defaults to `False`): Whether the model is used as a decoder or not. If `False`, the model is used as an encoder. use_cache (`bool`, *optional*, defaults to `True`): Whether or not the model should return the last key/values attentions (not used by all models). Only relevant if `config.is_decoder=True`. classifier_dropout (`float`, *optional*): The dropout ratio for the classification head. pre_norm (`bool`, *optional*, defaults to `False`): Whether to apply layer normalization before each block. adapter_reduction_factor (`int` or `float`, *optional*, defaults to 2): The factor by which the dimensionality of the adapter is reduced relative to `hidden_size`. adapter_layer_norm (`bool`, *optional*, defaults to `False`): Whether to apply a new layer normalization before the adapter modules (shared across all adapters). adapter_reuse_layer_norm (`bool`, *optional*, defaults to `True`): Whether to reuse the second layer normalization and apply it before the adapter modules as well. ln_before_adapter (`bool`, *optional*, defaults to `True`): Whether to apply the layer normalization before the residual connection around the adapter module. languages (`Iterable[str]`, *optional*, defaults to `["en_XX"]`): An iterable of language codes for which adapter modules should be initialized. default_language (`str`, *optional*): Language code of a default language. It will be assumed that the input is in this language if no language codes are explicitly passed to the forward method. Examples: ```python >>> from transformers import XmodConfig, XmodModel >>> # Initializing an X-MOD facebook/xmod-base style configuration >>> configuration = XmodConfig() >>> # Initializing a model (with random weights) from the facebook/xmod-base style configuration >>> model = XmodModel(configuration) >>> # Accessing the model configuration >>> configuration = model.config ```xmodc rt|d| ||d|||_||_||_||_||_||_||_||_ | |_ | |_ | |_ | |_ ||_||_||_||_||_||_||_||_t-||_||_y)N) pad_token_id bos_token_id eos_token_id)super__init__ vocab_size hidden_sizenum_hidden_layersnum_attention_heads hidden_actintermediate_sizehidden_dropout_probattention_probs_dropout_probmax_position_embeddingstype_vocab_sizeinitializer_rangelayer_norm_epsposition_embedding_type use_cacheclassifier_dropoutpre_normadapter_reduction_factoradapter_layer_normadapter_reuse_layer_normln_before_adapterlist languagesdefault_language)selfrrrrrrrrrrrrr rrrr r!r"r#r$r%r&r(r)kwargs __class__s i/var/www/skyplay_api_hub/venv/lib/python3.12/site-packages/transformers/models/xmod/configuration_xmod.pyrzXmodConfig.__init__rs: sl\hslrs$&!2#6 $!2#6 ,H)'>$.!2,'>$""4  (@%"4(@%!2i 0)i:wi r/i gelu皙?r1ig{Gz?g-q=rr2absoluteTNFr2FTT)en_XXN)__name__ __module__ __qualname____doc__ model_typer __classcell__)r,s@r-r r shP dJ%( # *!" !%54141r.r c6eZdZedeeeeefffdZy)XmodOnnxConfigreturncZ|jdk(rdddd}nddd}td|fd|fgS) Nzmultiple-choicebatchchoicesequence)rr3r2)rr3 input_idsattention_mask)taskr)r* dynamic_axiss r-inputszXmodOnnxConfig.inputssG 99) )&8 CL&:6Ll+!<0   r.N)r6r7r8propertyrstrintrGrr.r-r=r=s.   WS#X%6 67    r.r=N)r9 collectionsrtypingrconfiguration_utilsronnxrutilsr get_loggerr6loggerr r=__all__rr.r-rSsV #3   H %I1!I1Z  Z   ) *r.